Analysis of air pollution data at a mixed source location using boosted regression trees

This paper explores the use of boosted regression trees to draw inferences concerning the source characteristics at a location of high source complexity. Models are developed for hourly concentrations of nitrogen oxides (NO_X) close to a large international airport. Model development is discussed and methods to quantify model uncertainties developed. It is shown that good explanatory models can be developed and further, allowing for interactions between model variables significantly improves the model fits compared with non-interacting models. Methods are used to determine which variables exert most influence over predicted concentrations and to explore the NO_X dependency for each. Model predictions are used to estimate aircraft take-off contributions to total concentrations of NO_X and determine how these predictions are affected by annual variations in meteorological conditions and runway use patterns. Furthermore, the results relating to the aircraft contributions to total NO_X concentration are compared with those from a more detailed independent field campaign. Finally, we find empirical evidence that plumes from larger aircraft disperse more rapidly from the point of release compared with smaller aircraft. The reasons for this behaviour and the implications are discussed.     

Near roadway concentrations of organic source markers

Concentrations of elemental carbon (EC), organic carbon (OC), and 33 organic source markers (12 alkanes, 18 polycyclic aromatic hydrocarbons and ketones, and 3 hopanes) are reported near a highway in Raleigh, NC with an annual average daily traffic count of approximately 125,000 vehicles. Daily samples (24-h) were collected at two locations, one approximately 10 m and the other 275 m perpendicular from the road. Concentrations of OC were similar between near (mean = 7.6 μg m^?3) and far (8.0 μg m^?3) locations, but concentrations of most organic species at the near site were between 1.5 and 2 times higher than those at the far site.     

Investigations of an intense aerosol loading during 2007 cyclone SIDR – A study using satellite data and ground measurements over Indian region

Tropical cyclones are prominent weather systems characterized by high atmospheric pressure gradients and wind speeds. Intense tropical cyclones occur in India during the pre-monsoon (spring), early monsoon (early summer), or post-monsoon (fall) periods. Originating in both the Bay of Bengal (BoB) and the Arabian Sea (AS), these tropical cyclones often attain velocities of more than 100 km h^?1 and are notorious for causing intense rain and storm surge as they cross the Indian coast. In this study, we examine the changes in the aerosol properties associated with an intense tropical cyclone “SIDR”, that occurred during 11–16 November 2007 over BoB. This cyclone, accompanied with very strong surface winds reaching 223 km h^?1, caused extensive damage over Bangladesh. Ground-based measurements of Aerosol Optical Depth (AOD) in the neighboring urban environment of Hyderabad, India, showed significant variations due to changes in wind velocity and direction associated with the cyclone passage. The Terra-MODIS and AVHRR satellite images showed prevalence of dust particles mixed with emissions from anthropogenic sources and biomass-burning AS, while the aerosol loading over BoB was significantly lower. The positive values of Aerosol index (AI) obtained from the Ozone Monitoring Instrument (OMI) suggested the presence of an elevated aerosol layer over the West coast of India, AS and Thar Desert during and after the cyclone episode. Meteorological parameters from the MM5 mesoscale model were used to study the variations in winds associated with the cyclonic activity. Particulate matter loading over the region during the cyclone period increased by ～45% with an accompanying decrease in columnar aerosol optical depth. The variations in Angstrom parameters suggested coarse-mode particle loading due to dust aerosols as observed in satellite data.     

Air pollutant concentrations near three Texas roadways,Part I: Ultrafine particles

Vehicular emitted air pollutant concentrations were studied near three types of roadways in Austin, Texas: (1) State Highway 71 (SH-71), a heavily traveled arterial highway dominated by passenger vehicles; (2) Interstate 35 (I-35), a limited access highway north of Austin in Georgetown; and (3) Farm to Market Road 973 (FM-973), a heavily traveled surface roadway dominated by truck traffic. Air pollutants examined include carbon monoxide (CO), oxides of nitrogen (NO_x), and carbonyl species in the gas-phase. In the particle phase, ultrafine particle (UFP) concentrations (diameter < 100 nm), fine particulate matter (PM_2.5, diameter < 2.5 μm) mass and carbon content and several particle-bound organics were examined. All roadways had an upwind stationary sampling location, one or two fixed downwind sample locations and a mobile monitoring platform that characterized pollutant concentrations fall-off with increased distance from the roadways. Data reported in this paper focus on UFP while other pollutants and near-roadway chemical processes are examined in a companion paper. Traffic volume, especially heavy-duty traffic, wind speed, and proximity to the road were found to be the most important factors determining UFP concentrations near the roadways. Since wind directions were not consistent during the sampling periods, distances along wind trajectories from the roadway to the sampling points were used to study the decay characteristics of UFPs. Under perpendicular wind conditions, for all studied roadway types, particle number concentrations increased dramatically moving from the upwind side to the downwind side. The elevated particle number concentrations decay exponentially with increasing distances from the roadway with sharp concentration gradients observed within 100–150 m, similar to previously reported studies. A single exponential decay curve was found to fit the data collected from all three roadways very well under perpendicular wind conditions. No consistent pattern was observed for UFPs under parallel wind conditions. However, regardless of wind conditions, particle concentrations returned to background levels within a few hundred meters of the roadway. Within measured UFP size ranges, smaller particles (6–25 nm) decayed faster than larger ones (100–300 nm). Similar decay rates were observed among UFP number, surface, and volume.     

Heterogeneous uptake of octamethylcyclotetrasiloxane (D4) and decamethylcyclopentasiloxane (D5) onto mineral dust aerosol under variable RH conditions

We have carried out kinetic studies to characterize the heterogeneous decay of octamethylcyclotetrasiloxane (D₄) and decamethylcyclopentasiloxane (D₅) in the presence of representative mineral dust aerosol in order to obtain a better understanding of the atmospheric fate of these siloxanes. The heterogeneous chemistry of D₄ and D₅ with various mineral dusts was studied in an environmental aerosol reaction chamber using FTIR absorption spectroscopy to monitor the reaction. The apparent heterogeneous uptake coefficient, γ_app, for D₄ and D₅ with various mineral dusts was measured under dry conditions and as a function of relative humidity (RH). In addition, the effect of initial D₄ and D₅ concentration on the rate and yield of the reaction was examined. The uptake coefficient, γ_app, for D₄ and D₅ was similar for the most reactive aerosols tested, with kaolinite ≈hematite > silica. Limited uptake onto carbon black and calcite surfaces was observed for either siloxane. Reaction with hematite and kaolinite resulted in multilayer coverages, suggesting extensive polymerization of D₄ and D₅ on the aerosol surface.     

Temperature dependence of secondary organic aerosol

Temperature was found to have a dramatic effect on secondary organic aerosol formation from two ozonolysis systems, cyclohexene and α-pinene. Isothermal experiments were conducted for both systems where the lowest temperature, 278 K, formed approximately 2.5–3 times and 5–6 times the SOA formed at 300 K and 318 K, respectively. Changing the cyclohexene system temperature to a different isothermal experimental set point after completion of SOA formation did not lead to sufficient condensation/evaporation to reproduce the SOA formation at other temperature set points. When the system temperature was cycled between two set points at the end of an experiment, the α-pinene system showed reversibility between the initial temperature 318 K and 300 K. For temperature cycles between the initial temperature of 300 K–318 K, an irreversible loss of mass is observed after the first heating cycle with reversibility observed between subsequent temperature cycles. The SOA formed at 278 K was reversible over a 22 K range but was unable to evaporate sufficiently to match the SOA mass formed at 300 K. Hygroscopicity measurements, taken after the completion of SOA formation, indicate that hygroscopicity of the aerosol is also a function of temperature and that the aerosol does not continue to be oxidized after initial growth is complete. The differing hygroscopicity of the semi-volatile component of the aerosol is evident during system temperature changes after completion of the experiment.     

Effects of post-sampling conditions on ambient carbon aerosol filter measurements

Ambient carbonaceous material collected on quartz filters is prone to measurement artifacts due to material gained or lost during post-sampling field latency, shipping, and storage. In seventeen sampling events over a one year period, ambient PM_2.5 aerosols were collected on quartz filters (without denuders) and subjected to various filter treatments to assess the potential for and extent of artifacts. The filter treatments simulated post-sampling environments that filters may be exposed to and included: storage at 40 °C for up to 96 h, storage at ?16 °C for 48 h, and storage at room temperature (～21 °C) for 48 h. Carbon mass on the filters was measured using a thermal-optical method. The total carbon (TC), total organic carbon (TOC) and total elemental carbon (TEC) as well as carbon thermal fraction masses were obtained. Statistical analyses were performed to identify significant differences in carbon fraction concentrations between filters analyzed immediately after sampling and after being subjected to treatment.TOC and TC concentrations decreased by on average 15 ± 5% and 10 ± 4%, respectively, for filters maintained at 40 °C for 96 h but did not change for filters stored at room temperature or frozen for 48 h. TEC did not change for any of the filter treatments. The mass concentration for the organic carbon thermal fraction that evolves at the lowest temperature step (OC1) decreased with increasing storage time at 40 °C with average losses of 70 ± 7% after 96 h. Therefore, OC1 is not a stable measurement due to post-sampling conditions that may be encountered. This work demonstrates that TOC and TC can have substantial measurement artifacts on filters subjected to field latency and other non-temperature controlled post-sampling handling, compared to the carbon loadings on the filter at the end of the sampling period.     

Predictive models of turbidity and water depth in the Doñana marshes using Landsat TM and ETM+ images

We have used Landsat-5 TM and Landsat-7 ETM+ images together with simultaneous ground-truth data at sample points in the Doñana marshes to predict water turbidity and depth from band reflectance using Generalized Additive Models. We have point samples for 12 different dates simultaneous with 7 Landsat-5 and 5 Landsat-7 overpasses. The best model for water turbidity in the marsh explained 38% of variance in ground-truth data and included as predictors band 3 (630–690 nm), band 5 (1550–1750 nm) and the ratio between bands 1 (450–520 nm) and 4 (760–900 nm). Water turbidity is easier to predict for water bodies like the Guadalquivir River and artificial ponds that are deep and not affected by bottom soil reflectance and aquatic vegetation. For the latter, a simple model using band 3 reflectance explains 78.6% of the variance. Water depth is easier to predict than turbidity. The best model for water depth in the marsh explains 78% of the variance and includes as predictors band 1, band 5, the ratio between band 2 (520–600 nm) and band 4, and bottom soil reflectance in band 4 in September, when the marsh is dry. The water turbidity and water depth models have been developed in order to reconstruct historical changes in Doñana wetlands during the last 30 years using the Landsat satellite images time series.     

Evaluation of the aggressive potential of marine chloride and sulfate salts on mortars applied as renders in the Metropolitan Region of Salvador--Bahia, Brazil

In recent years, growing interest has focused on determining the performance of materials and evaluating the service life of structures exposed to various environmental forces. In this context, the determination of the aggressive potential of marine salts on mortars used as external renders is critical. The present study aimed to evaluate the spatial distribution of marine salts relative to distance from the sea. This was done by monitoring the deposition rate of chlorides and sulfates in wet candle sensors, located at nine stations scattered around the Metropolitan Region of Salvador, state of Bahia, Brazil. The study also determined the effectiveness of water-soluble salts at penetrating three different types of mortars of varying cement content via deposition and diffusion. The methodology employed enabled an evaluation of the efficiency of the monitoring sensors' measurement of the aggressiveness potential of local marine aerosol, and determination of the comparative performance of the three mortars tested, from the standpoint of resistance to salt penetration. The type and amount of salts captured both in solution and in powder samples extracted from the mortars were determined by ion chromatography. The analysis of the various types of mortars tested indicated which types are more resistant to the aggressive potential of the region's marine aerosol and the distance from the shore where local buildings are liable to be most strongly affected.     

Global warming and environmental production efficiency ranking of the Kyoto Protocol nations

This paper analyzes the United Nations Organization's Kyoto Protocol nations to address two questions. First, what are the environmental production efficiency rankings of these nations? Second, is there a relationship between a nation's ratification status and its environmental production efficiency ranking? Our findings suggest that the nations that have ratified the Kyoto Protocol are more likely to be environmentally production efficient as compared to the nations that have not ratified the Protocol.